CodeImprove: Program Adaptation for Deep Code Models

TL;DR

CodeImprove introduces a novel approach to adapt out-of-scope code inputs for deep learning models by using validation metrics and genetic algorithms, significantly improving model accuracy and input validation effectiveness.

Contribution

This paper presents a new method for program input adaptation that enhances deep code model performance without frequent retraining, using program transformation and validation techniques.

Findings

Up to 8.78% accuracy improvement in code models.

51.28% relative improvement in handling out-of-scope inputs.

High effectiveness in detecting out-of-scope inputs with AUC of 0.924.

Abstract

Leveraging deep learning (DL)-based code analysis tools to solve software engineering tasks is becoming increasingly popular. Code models often suffer performance degradation due to various reasons (e.g., code data shifts). Retraining is often required to address these issues, but frequent model updates are costly in labeling and deployment. In this paper, we explore an alternative solution: Adapting the program inputs to the code models. This can be achieved by two steps: 1) input validation that focuses on identifying whether an input is an out-of-scope input program that are beyond a model's handling capability, and 2) input adaptation that adapts out-of-scope inputs to become in-scope inputs. Validating program input is challenging, as current techniques focus on continuous inputs such as image data and fail with discrete inputs like code data, which have unique characteristics and are processed differently by deep learning models. Adapting out-of-scope programs is also challenging due to their vast search spaces. Therefore, in this paper, we propose CodeImprove, which distinguishes out-of-scope from normal inputs and converts such out-of-scope inputs back to in-scope inputs through program transformation. In particular, we propose a validity score metric to identify out-of-scope inputs and leverage genetic algorithms to apply semantic preserving program transformation to convert out-of-scope inputs to in-scope inputs. Our experimental results show CodeImprove can enhance up to 8.78% of accuracy, and 51.28% of relative improvements in three code models on two SE tasks. Additionally, our input validation is promising in detecting out-of-scope inputs (AUC score of 0.924).